Quantitative analysis of brain structure is important in the study of many neurological
and neuropsychiatric disorders. Furthermore, the accurate delineation of structure can
provide important baseline information for quantifying brain function. This talk will
present a body of work grounded in the use of geometrical constraints and mathematical
optimization to analyze neuroanatomical structure of the human brain from Magnetic
Resonance Images (MRI).
More specifically, we will describe our applied mathematical approaches to: the
segmentation of cortical and subcortical structure, the analysis of white matter fiber
tracks using diffusion tensor imaging (DTI) and the intersubject registration
of neuroanatomical (aMRI) datasets. Many of our methods rally around the use of
geometric constraints and level set evolution strategies. In addition, we are currently
merging the methods described here with functional MRI (fMRI) analysis strategies with
the goal of developing a more fully integrated approach to functional/structural brain
image analysis. The analysis of gray matter structure and connecting white matter paths
combined with the ability to bring all information into a common space via intersubject
registration should provide us with a rich set of data to investigate structure and variation
in the (normal, abnormal and developing) human brain, as well as provide a basis
for current work in the development of integrated brain function/structure analysis.
In the accompanying figure we show an example of our efforts related to segmenting the
amygdala and the hippocampus from T1-weighted MRI data. These subcortical
structures are important in the study of autism and other disorders. The approach is
based on the use of a 3D level set parametrization of the object surfaces and uses interobject
statistical priors to constrain the solution. The left to right sequence shows three
stages of the simultaneous evolution of the four (1 each for the left and right
hippocampus and left and right amygdala) different zero level sets both overlaid on the
image data and as a 3D visualization.
Joint work with Xenios Papademetris, Jing Yang, Marcel Jackowski,
Xiaolan Zeng and Lawrence Staib (Yale
University, Biomedical Engineering, Diagnostic Radiology and Electrical Engineering).
Copyright. All Rights Reserved.